תעתיק 🔗
WEBVTT
00:00:09.839 --> 00:00:16.320
Hello, you're listening to EPITalk: Behind the Paper, a monthly podcast from the annals of epidemiology.
00:00:16.320 --> 00:00:26.399
I'm Patrick Sullivan, editor-in-chief of the journal, and in this series we take you behind the scenes of some of the latest epidemiologic research featured in our journal.
00:00:26.399 --> 00:00:38.000
Hello, you're listening to Epitalk Behind the Paper, a monthly podcast from the Annals of Epidemiology.
00:00:38.000 --> 00:00:47.039
I'm Patrick Sullivan, editor-in-chief of the journal, and in this series, we take you behind the scenes of some of the latest epidemiologic research featured in our journal.
00:00:47.039 --> 00:00:48.560
Today we're here with Dr.
00:00:48.560 --> 00:01:03.920
Derek Ng to discuss his article, A Comparison of Neural Networks and Regression-Based Approaches for Estimating Kidney Function in Pediatric Chronic Kidney Disease, Practical Predictive Epidemiology for Clinical Management of a Progressive Disease.
00:01:03.920 --> 00:01:13.120
You can read the full article online in the May 2025 issue of the journal at www.annalsof epidemiology.org.
00:01:13.120 --> 00:01:13.599
Dr.
00:01:13.599 --> 00:01:25.840
Derek Ng is an associate professor of epidemiology at the Johns Hopkins Bloomberg School of Public Health and is the principal investigator of the Data Coordinating Center for the Chronic Kidney Disease and Children Cohort Study.
00:01:25.840 --> 00:01:34.079
His research focuses on pediatric epidemiology with a particular focus on pediatric nephrology and pediatric critical care.
00:01:34.079 --> 00:01:44.879
His methodologic interests are in the multi-center cohort study designs, prediction modeling, and translational epidemiologic tools to help clinicians, patients, and families.
00:01:44.879 --> 00:01:45.439
Dr.
00:01:45.439 --> 00:01:47.439
Ng, thanks so much for joining us today.
00:01:47.760 --> 00:01:49.120
Thank you for having me.
00:01:49.760 --> 00:02:01.120
I'd like to start out by just asking if you can share with us some background on the problem described in your paper, both pediatric kidney disease and then the modeling question that your study addressed.
00:02:01.120 --> 00:02:02.640
Why are these issues important?
00:02:03.040 --> 00:02:03.280
Sure.
00:02:03.280 --> 00:02:09.280
So children with chronic kidney disease represent a very important clinical population.
00:02:09.280 --> 00:02:12.639
This is a rare disease, but it's not necessarily uncommon.
00:02:12.639 --> 00:02:16.080
We actually don't know exactly how many children in the U.S.
00:02:16.080 --> 00:02:23.360
are afflicted with chronic kidney disease, mainly because some are asymptomatic and others are very severe.
00:02:23.360 --> 00:02:29.039
And the main clinical concern are those who have serious disease.
00:02:29.039 --> 00:02:47.120
And what we aim to do in epidemiology is to help clinicians and patients and their families try to make sense of this disease, try to
Load full transcript
sort through what the best treatment options are, and also to understand how significant the disease is in terms of kidney function.
00:02:47.120 --> 00:02:53.439
One of the clinical concerns of pediatric nephrology is how well is the kidney functioning?
00:02:53.439 --> 00:02:55.280
There are ways to measure this.
00:02:55.280 --> 00:03:15.360
These procedures are long between five and six hours and require the injection of a contrast, but then there are ways to try to predict kidney function or estimate kidney function by linking biomarkers that we know are associated with kidney function to the directly measured quantities of kidney function.
00:03:15.360 --> 00:03:26.400
So that we can use mathematical and statistical models in order to estimate or predict what the kidney function is without having to measure it directly.
00:03:26.400 --> 00:03:38.159
The main benefit of this is that biomarkers measured in a simple blood draw can help clinicians understand how serious the kidney disease is in this patient population.
00:03:38.560 --> 00:03:44.000
And that sounds like say some procedures for the patients and the children as well.
00:03:44.400 --> 00:03:46.319
Yeah, that's a big benefit.
00:03:46.319 --> 00:03:48.560
It's much easier on the patients.
00:03:48.560 --> 00:03:50.879
It's quicker, it's also less expensive.
00:03:50.879 --> 00:04:02.879
So these blood draws are a much more efficient and cost-effective way to estimate kidney function as long as the models that we use to estimate kidney function are good.
00:04:03.360 --> 00:04:07.360
So I think that leads then to what the main sort of purpose of the study was.
00:04:07.360 --> 00:04:17.519
Can you say a little bit more about the different approaches that you were evaluating and why you felt like it was important to evaluate that to compare these two approaches to the estimation?
00:04:17.839 --> 00:04:18.160
Right.
00:04:18.160 --> 00:04:43.680
So we should say first that the first estimating equations for kidney function were published in 1976, but there was a real big breakthrough in adult clinical nephrology in 1999 when an equation was developed based on classical statistical regression methods to estimate kidney function based on a biomarker called serum creatinine.
00:04:43.680 --> 00:04:55.439
This was later updated and refined through different equations, but the structure of it remained the same in terms of using a regression-based approach to estimate kidney function.
00:04:55.439 --> 00:05:06.720
Since that time, there's also been a second biomarker that has been found to be very useful to estimate kidney function, as well as also being convenient to be measured clinically.
00:05:06.720 --> 00:05:13.839
This second biomarker is called cystatin C, and it's typically used as a sequential test.
00:05:13.839 --> 00:05:32.000
If there's something going on with the serum carotinine that looks suspicious, or the clinician would like kind of a second opinion from these estimating
equations, they could use serum cystatin C in order to get another estimate of kidney function to try to understand the severity of kidney disease.
00:05:32.000 --> 00:05:41.519
Now, when these equations were developed, neural networks and machine learning methods were not as well developed or accessible.
00:05:41.519 --> 00:05:51.120
I think probably most listeners understand that artificial intelligence, machine learning, has become more accessible and common over time.
00:05:51.120 --> 00:06:05.439
And so what we were really interested in was whether the neural network actually offered a better estimate under the clinical constraints of having two biomarkers measured, which would reflect the clinical conditions that physicians operate under.
00:06:05.439 --> 00:06:16.480
If we found that the neural network offered something new, this would be an important finding that we could use to estimate GFR and think about ways to translate that.
00:06:16.480 --> 00:06:24.079
However, if they didn't perform as well, it would also give the clinicians confidence in the tools that they currently have existing.
00:06:24.079 --> 00:06:37.839
But I think among the talk about the promise of artificial intelligence and machine learning, it represented an opportunity to explore whether this was a great benefit in the context of clinical nephrology.
00:06:38.399 --> 00:06:48.959
Yeah, and a really specific application of that question of like what can these kinds of new tools do better and where where are the existing systems maybe already robust?
00:06:48.959 --> 00:06:52.879
There may be listeners who are less familiar with neural networks.
00:06:52.879 --> 00:07:02.560
So just at a high level, could you give us an idea of what a neural network is in a epidemiologist's terms, but you know, why that technology might be helpful?
00:07:03.040 --> 00:07:03.360
Sure.
00:07:03.360 --> 00:07:07.600
So the main purpose of these neural networks is ultimately to predict some outcome.
00:07:07.600 --> 00:07:10.079
Now that outcome could be about classification.
00:07:10.079 --> 00:07:18.800
In this particular case, glomerular filtration rate or GFR, which is a measure of kidney function, is a continuous variable.
00:07:18.800 --> 00:07:30.480
And so the neural network is really well suited to try to predict this continuous variable of glomerular filtration rate based on predictors.
00:07:30.480 --> 00:07:35.600
Now, in a classical regression setting, the same goal can be achieved.
00:07:35.600 --> 00:07:56.639
I think the distinction is that the neural network integrates information in a nonlinear way, unlike that of regression, and does this using what are called hidden layers and hidden units in order to come up with a final prediction that is close to what we conceptualize as the truth.
00:07:57.040 --> 00:07:57.439
Great.
00:07:57.439 --> 00:08:04.639
So given a better understanding of the neural networks, just walk us through the methods that you use to kind of make this comparison.
00:08:04.959 --> 00:08:19.680
This work capitalized on a paper that we published in 2021 that offered a new system of equations for children and adolescents and young adults under the age of 25 to estimate their GFR.
00:08:19.680 --> 00:08:23.759
This was based on a classical regression structure.
00:08:23.759 --> 00:08:39.600
And we found that it offered, on average, unbiased, reliable, and highly correlated predictions of GFR compared to the directly measured GFR data that we had in this very large study of children with chronic kidney disease.
00:08:39.600 --> 00:08:46.960
What we intended to do in this paper was to use a neural network approach to estimate the same thing.
00:08:46.960 --> 00:09:02.559
And I think what was very unique about this paper was that we used the exact same training and testing data sets that were presented in the original 2021 paper in this new exploration of prediction methodology.
00:09:02.559 --> 00:09:08.960
And this way we could compare directly how good the agreement was between these two different approaches.
00:09:09.360 --> 00:09:14.240
And now you start to get into the language of epidemiologists around the agreement of these.
00:09:14.240 --> 00:09:31.279
So that even if the I think to probably many of us, myself included, like the functioning of the neural network might be a little bit of a black box, but the premise, which is you have like a gold standard method and you have a method that may offer benefits, and thinking about the agreement really sort of comes back into the epi wheelhouse.
00:09:31.279 --> 00:09:34.240
So having said that, what were some of the key findings?
00:09:34.639 --> 00:09:43.919
So what we found was that the neural network did not offer better predictions of GFR compared to the classical approach.
00:09:43.919 --> 00:09:47.200
Now, having said that, it also didn't do much worse.
00:09:47.200 --> 00:09:49.279
In fact, they were very comparable.
00:09:49.279 --> 00:09:55.440
And it's also fair to say that the neural network methodology was more complex.
00:09:55.440 --> 00:10:10.159
And the output, while we were able to predict a single value of GFR, there was a lot of architecture and complexity in the model that wasn't present in the classical regression-based setting.
00:10:10.159 --> 00:10:24.399
So, in the scenario where you have two predictions that yield similar agreement and are both doing quite well, if you have your choice between a simpler model and a more complex one, it's better to go with the simple model.
00:10:24.399 --> 00:10:33.519
And we thought carefully about what sort of translational implications there were, considering that they both performed equally well.
00:10:33.519 --> 00:10:41.679
And we hit upon some of the benefits of the simple model compared to the more complex neural network one.
00:10:42.320 --> 00:10:48.799
So your recommendation is that for now, the sort of simpler, like traditional model is probably preferable.
00:10:49.200
--> 00:10:52.480
Yeah, it it certainly seems to be sufficient.
00:10:52.480 --> 00:10:59.200
And I think that that is also because the number of predictors that we had were quite limited.
00:10:59.200 --> 00:11:04.320
We were limited to two biomarkers that are commonly measured clinically.
00:11:04.320 --> 00:11:15.360
And I think that the neural network probably has a better opportunity to shine and to do better if we had a lot more clinical predictors available.
00:11:15.360 --> 00:11:19.840
However, that's just not the case right now clinically, and hopefully that will change.
00:11:19.840 --> 00:11:21.600
It could be a while though.
00:11:21.600 --> 00:11:30.320
Right now, there's a lot of discovery work to try to identify new biomarkers that can be used to help estimate GFR.
00:11:30.320 --> 00:11:38.159
But right now, given the millions of blood tests that go on every day, there's only one, serum creatinine, that's measured routinely.
00:11:38.159 --> 00:11:41.759
And then within a small fraction of those is cystatin C.
00:11:41.759 --> 00:11:52.159
So the idea that we will be getting, you know, dozens of more clinical predictors specifically for kidney function in a blood test, I think it could be a while before that happens.
00:11:52.559 --> 00:12:12.559
So is it too broad of a conclusion to say that these kind of tools might be more useful where there are more inputs to the I think I'm I'm translating a little bit, but if you only have a couple inputs, the that the neural networks may not have much room to improve or look at those in exponential ways compared to where there's a richer set of predictors.
00:12:12.960 --> 00:12:14.639
I think that's a great way to put it.
00:12:14.639 --> 00:12:17.120
And I think that's what we're we're up against.
00:12:17.120 --> 00:12:28.960
And as I mentioned before, you know, one of the motivations was that speaking to clinicians and physicians, nurses, and hearing about, well, hey, have you tried a neural network?
00:12:28.960 --> 00:12:31.840
Have you tried machine learning to try to predict GFR?
00:12:31.840 --> 00:12:35.440
Maybe there's something there that we're missing currently if you haven't done it.
00:12:35.440 --> 00:12:40.480
And we certainly kept an open mind about it and tried to do the best that we could.
00:12:40.480 --> 00:12:47.519
But I think as as you mentioned, the limiting factor is the fact that we only have two biomarkers.
00:12:47.519 --> 00:12:55.600
Now, these two biomarkers are very, very good, but they are only two, and these neural networks are really designed for many, many predictors.
00:12:55.600 --> 00:13:06.559
We did try to allow the neural network more of a chance to deal with more predictors by using transformations of the original predictors.
00:13:06.559 --> 00:13:13.279
So including polynomials for the serum caratinine, interactions with age as a predictor.
00:13:13.279 --> 00:13:21.120
And even with that, we found that we couldn't do any better than the original under 25 equations.
00:13:21.679 --> 00:13:29.679
So I want to turn now to a segment of the podcast that gets a little bit more into how you came to this work and your career.
00:13:29.679 --> 00:13:36.320
I think it's always an opportunity to sort of learn from how people get to where they are to doing really interesting work.
00:13:36.320 --> 00:13:43.759
And I'll speak for myself as a later career, you know, researcher to try and share, you know, things that you may have learned along the way.
00:13:43.759 --> 00:13:52.080
So maybe just starting about like the inspiration behind the paper, or maybe more broadly, your larger focus in pediatric nephrology.
00:13:52.080 --> 00:13:55.600
How did that grow as a passion or a research area for you?
00:13:55.919 --> 00:13:57.279
It's a great question.
00:13:57.279 --> 00:14:02.799
So I trained at Hopkins and started with a Masters of Science in Epidemiology.
00:14:02.799 --> 00:14:07.600
And I was fortunate to make a nice connection with one of the faculty.
00:14:07.600 --> 00:14:09.440
And his name was Dr.
00:14:09.440 --> 00:14:15.279
Alvaro Muñoz, and he has been a great mentor of mine for many, many years.
00:14:15.279 --> 00:14:22.879
And at the time, though, when I had just met him, we discussed one of the papers I wrote for his class, and he was interested in that.
00:14:22.879 --> 00:14:25.440
And I was looking for a job at the time.
00:14:25.440 --> 00:14:26.879
I wasn't sure what I was going to do.
00:14:26.879 --> 00:14:41.360
And he kindly said, "Look, if you're interested in a job, I think we have a job opening coming up." And the job was to do programming and epidemiologic analysis in this cohort of children with chronic kidney disease.
00:14:41.360 --> 00:14:47.519
This was back in 2009, and the study had just started in 2003.
00:14:47.519 --> 00:14:53.679
So it was accumulating data, and there was a lot of opportunity, and I was really grateful for the opportunity.
00:14:53.679 --> 00:15:06.000
And since then, I've spent my whole career working largely, but not exclusively, in the domain of observational cohort study of chronic kidney disease in children.
00:15:06.000 --> 00:15:08.320
I can't say enough about this study.
00:15:08.320 --> 00:15:29.279
It's a beautiful national treasure that's a testament to the work of the participants, the families, the clinical coordinators, the people who measure samples at our central biochemistry laboratory, and just this beautiful team effort to try to come up with answers for this population that needs help.
00:15:29.279 --> 00:15:37.360
And it's hard not to understand the importance and imperative to help children who are sick.
00:15:37.360 --> 00:15:41.200
And that's really the primary mission of this cohort study.
00:15:41.200 --> 00:16:02.320
So I view my role as coming up with answers to questions that are formulated in a collaborative way and working towards offering knowledge so that the physicians can treat children, help them
and their families, and give them the best information so that the children can live their best lives.
00:16:03.039 --> 00:16:03.919
Yeah, thanks for that.
00:16:03.919 --> 00:16:06.320
And I think it's an especially important time.
00:16:06.320 --> 00:16:21.440
Like the what you share is really important because it reminds us that one, uh wherever we are in the mentor, you know, lineage here, and you go from being, you know, having a mentor give you a step up to becoming a mentor and giving other people a step up.
00:16:21.440 --> 00:16:30.240
But also just that that the arc of science is long and and that it's hard to predict at that, like when you entered into this, that this would be what you were doing.
00:16:30.240 --> 00:16:37.759
But to me, that's sort of a really timely topic right now, because there are threats to these pipelines of research.
00:16:37.759 --> 00:16:56.240
And one of the things I think you really only come to fully appreciate by living in our in this world is that things are nonlinear and that these kind of cohort studies, while they're observational, are such a rich source for derivative science or second gen science that moves things along.
00:16:56.240 --> 00:16:59.679
And you just don't know where the discovery is going to come from.
00:16:59.679 --> 00:17:14.720
So I one, this is a great example of like NIH, presumably as a funder, like making an investment in something that both supports careers of earlier career people when you were earlier career, and that gives rise to observations that we can't always anticipate.
00:17:14.720 --> 00:17:21.200
And we're we're living in a time where there's a fair amount of transactionality, I think, in terms of the research interests and the investments.
00:17:21.200 --> 00:17:32.640
And um, this is just a useful example of how when you plant good seeds in good ground and you give them good water, you know, things are going to grow and it may go beyond what you know what you plan.
00:17:33.359 --> 00:17:35.279
Yeah, I think that's a great way to put it.
00:17:35.279 --> 00:17:54.640
And I think often about in my field particular, defending our research portfolio and the value of this sort of research, especially because something like estimating kidney function is not something that you might consider should be something very profitable, at least monetarily.
00:17:54.640 --> 00:18:01.359
I don't see that measuring kidney function should be something that uh is monetarily profitable.
00:18:01.359 --> 00:18:13.680
Um, instead, this is simply providing information to clinicians so that they can treat children with kidney disease properly and make the best decisions and understand more about it.
00:18:13.680 --> 00:18:35.440
And of course, this uh cohort study, which is generously funded by NIH and has been since the inception, has yielded a lot of important information, not just about kidney function, but about the natural history of kidney disease in children, which until the study was started, not much had been known.
00:18:35.440 --> 00:18:41.039
And the amount of information that's been generated from this cohort study has really been remarkable.
00:18:41.039 --> 00:18:48.000
I think there's been over 200 papers and a lot of investigators working on it, and it really is a national resource.
00:18:48.000 --> 00:18:56.079
So people who are interested in using this type of data are able to get it at the NIDDK repository online.
00:18:57.519 --> 00:18:58.160
That's great.
00:18:58.160 --> 00:19:04.960
And I'll get a URL from you and we'll post that up when we announce the podcast in the show notes as well, so people can access it.
00:19:04.960 --> 00:19:14.559
So you, I mean, we are we're talking a little bit about, um, and I appreciate you sharing about your kind of the leg up that you got from like a later career colleague.
00:19:14.559 --> 00:19:21.359
Now that you're at a different place in your career, what advice would you give to young researchers or clinicians who are interested in this field?
00:19:21.359 --> 00:19:28.079
I imagine a lot of our earlier career colleagues are seeing this these kinds of tools as an important part of the future of science.
00:19:28.079 --> 00:19:35.359
So, what advice do you have for earlier career people who were thinking about wading into this new pool of epidemiology and data science?
00:19:35.759 --> 00:19:42.480
I think the main thing is having an openness to learn and explore new ideas.
00:19:42.480 --> 00:19:47.200
And I have to confess, learning about the neural networks was actually very difficult.
00:19:47.200 --> 00:19:53.759
There were two great resources available to help make sense of this and understand it.
00:19:53.759 --> 00:19:59.119
And that is a book called The Introduction to Statistical Learning, the second edition.
00:19:59.119 --> 00:20:01.519
It was Invaluable.
00:20:01.519 --> 00:20:09.599
And my team and I read this as kind of like a book club in order to understand new methods and it walked through everything quite well.
00:20:09.599 --> 00:20:26.000
But what was nice was that we even though we weren't enrolled in a class or trying to get a degree or anything like that, we did this uh on our own and learned together as a team, a group of about uh six faculty and biostatisticians and epidemiologists.
00:20:26.000 --> 00:20:27.759
We could learn together and talk about it.
00:20:27.759 --> 00:20:30.880
And I think that's something very special about an academic environment.
00:20:30.880 --> 00:20:32.880
So don't be afraid to learn new things.
00:20:32.880 --> 00:20:43.519
The other thing I would say is that it's really important for epidemiologists to listen to clinicians and physicians and to understand what they need.
00:20:43.519 --> 00:20:50.319
I don't think that epidemiology just for the sake of Epi is always the best.
00:20:50.319 --> 00:20:52.880
And ultimately what we're trying to do is to help people.
00:20:52.880 --> 00:20:59.440
So if we can
learn from the clinicians and physicians, you know, what frustrates them?
00:20:59.440 --> 00:21:01.039
What are they uncertain about?
00:21:01.039 --> 00:21:05.039
What would they like to know more about in order to treat their patients better?
00:21:05.039 --> 00:21:20.559
These are the sorts of questions that can help improve our own research as epidemiologists, because we don't want it simply to be in the ivory tower and not translate, um, but we want to share it with the people who will use it.
00:21:20.960 --> 00:21:21.920
That's great advice.
00:21:21.920 --> 00:21:33.200
And I think that the collaboration between epidemiologists and clinicians, I put myself in the epidemiologist camp, and it just helps with the relevance of the questions that we're asking and the impact.
00:21:33.200 --> 00:21:40.720
And I think, in fairness to us, to our part of that partnership, like we really have no way to know what those questions are unless we ask.
00:21:40.720 --> 00:21:44.880
So I think it's a really smart way and it increases the significance of what we do.
00:21:44.880 --> 00:21:45.920
So all right.
00:21:45.920 --> 00:21:46.319
Dr.
00:21:46.319 --> 00:21:49.519
Ang, do you have any last thoughts you'd like to share with our listeners?
00:21:49.839 --> 00:21:59.119
What I'd like to implore, any listeners and students, I do teach a lot of students here, and I encourage them to try to answer big questions.
00:21:59.119 --> 00:22:07.599
Uh, think big, go out and collect data, collect new data, things that are not easily found, because that's where the discovery happens.
00:22:07.599 --> 00:22:15.680
I think that in general, it can be very easy to find convenient sources of data, and that can be very useful as a starting point.
00:22:15.680 --> 00:22:19.519
But as epidemiologists, we're often concerned about selection bias.
00:22:19.519 --> 00:22:27.839
We might have to get out of our comfort zone and pursue avenues that will yield new data, data that hasn't been measured yet.
00:22:27.839 --> 00:22:31.839
And that tends to be more valuable data.
00:22:31.839 --> 00:22:39.759
And I think that that's something I try to impart on my students, and hopefully listeners can also benefit from that.
00:22:40.720 --> 00:23:02.720
Yeah, it is it is interesting that I think sometimes in training, like during a master's and PhD programs, there is a little bit of a bias to look for data sources that are well understood because we want the predictability, you know, for our colleagues who are in training, of not diving into a pool where you can't see the bottom, you know, that it's an answerable question and that the data are likely to be able to use to answer it.
00:23:02.720 --> 00:23:07.519
But the highest yield science is um, in some regards, the higher risk science.
00:23:07.519 --> 00:23:15.039
And so I think trying to strike that right balance between like predictability for someone, because the goal is to graduate, you know, the goal is to get the skills and graduate.
00:23:15.039 --> 00:23:19.599
But I think at every stage of career, there's some incentive to take the safer route.
00:23:19.599 --> 00:23:30.319
So I think your your message is really important, one, which is how far can we push into doing something that may be higher risk, higher reward, and when is the right time in one's career to do that?
00:23:30.319 --> 00:23:33.839
And probably every time in your career is the right time to do that.
00:23:33.839 --> 00:23:38.480
But there are then there are all these kind of other structures and incentives towards um predictability.
00:23:38.480 --> 00:23:43.680
So it's a complicated call as an academic researcher and as someone who's in training as well.
00:23:44.000 --> 00:23:44.559
Absolutely.
00:23:44.559 --> 00:23:52.880
And I think there's certainly uh time and place, and particularly training, you know, to uh make best use of time and and still have impactful science.
00:23:52.880 --> 00:23:55.440
And um, there's certainly wonderful resources.
00:23:55.440 --> 00:24:04.000
I guess my point is that in order to defend our space well, it means that we have to bring something new to the table.
00:24:04.000 --> 00:24:11.839
And as you said, sometimes the the higher risk data collection can be higher reward, and hopefully it is.
00:24:11.839 --> 00:24:14.240
But at the very least, we shouldn't be scared of it.
00:24:14.240 --> 00:24:22.160
I mean, if there are appropriate opportunities to use existing data, that's uh fantastic and appropriate is best.
00:24:22.160 --> 00:24:29.920
But we should be open about what we don't know and maybe consider taking a difficult route to collect new things.
00:24:30.240 --> 00:24:41.200
And one of the things I really like about your work is that it links up kind of a new technology that can feel a little black boxy or a little intimidating with some more traditional methods.
00:24:41.200 --> 00:24:49.839
And I think like that those are uh places where epidemiologists should be leading as people who like to hang out with data and make friends with data.
00:24:49.839 --> 00:24:53.119
This is the right kind of question for our field, I think.
00:24:53.119 --> 00:25:00.319
And so, you know, I appreciate your work and I hope that we see more of this coming into the journal using those epi methods and some newer techniques.
00:25:00.640 --> 00:25:01.839
I really appreciate that.
00:25:01.839 --> 00:25:10.559
And I have to say, my mentor deplored black boxes and insisted that we demystify everything.
00:25:10.559 --> 00:25:15.359
And if we can't explain it and can't write it out, then we don't understand it.
00:25:15.359 --> 00:25:19.440
And our whole point of being in this field is for understanding and knowledge.
00:25:19.440 --> 00:25:21.359
So I really appreciate that.
00:25:21.359 --> 00:25:28.880
I'm so thankful for the opportunity to talk more about this work and the opportunity to have this work published in Annals of Epidemiology.
00:25:29.680 --> 00:25:31.920
Well, thank you for sending the work to us.
00:25:31.920 --> 00:25:34.960
Thank you for taking the time to be on this episode.
00:25:34.960 --> 00:25:36.799
Really appreciate you joining us today.
00:25:36.799 --> 00:25:39.200
It was such a pleasure to have a chance to talk with you.
00:25:39.200 --> 00:25:40.559
Thanks so much.
00:25:40.559 --> 00:25:42.960
I'm your host, Patrick Sullivan.
00:25:42.960 --> 00:25:47.279
Thanks for tuning in to this episode and see you next time on EPITalk.
00:25:47.279 --> 00:25:53.279
Brought to you by Annals of Epidemiology, the official journal of the American College of Epidemiology.
00:25:53.279 --> 00:26:06.160
For a transcript of this podcast or to read the article featured on this episode and more from the journal, you can visit us online at www.annals of epidemiology.org..